HAL Id: hal-01771622
https://hal-univ-rennes1.archives-ouvertes.fr/hal-01771622
Submitted on 5 Jul 2018
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Locoregional treatments before liver transplantation for hepatocellular carcinoma a study from the European
Liver Transplant Registry
Hans-Christian Pommergaard, Andreas Arendtsen Rostved, René Adam, Lau Caspar Thygesen, Mauro Salizzoni, Miguel Angel Gómez Bravo, Daniel Cherqui, Paolo de Simone, Karim Boudjema, Vincenzo Mazzaferro, et al.
To cite this version:
Hans-Christian Pommergaard, Andreas Arendtsen Rostved, René Adam, Lau Caspar Thygesen, Mauro Salizzoni, et al.. Locoregional treatments before liver transplantation for hepatocellular carci- noma a study from the European Liver Transplant Registry. Transplant International, Wiley, 2018, 31 (5), pp.531-539. �10.1111/tri.13123�. �hal-01771622�
Accepted Article
DR. HANS-CHRISTIAN POMMERGAARD (Orcid ID : 0000-0001-7068-1946) DR. ANDREAS ARENDTSEN ROSTVED (Orcid ID : 0000-0002-3359-7734)
Article type : Original Article
Locoregional treatments before liver transplantation for
hepatocellular carcinoma: A study from the European Liver Transplant Registry
Authors
Hans-Christian Pommergaard1 MD PhD, Andreas Arendtsen Rostved1 MD, René Adam2 MD Professor, Lau Caspar Thygesen3 Associate professor PhD, Mauro Salizzoni4 MD Professor, Miguel Angel Gómez Bravo5 MD Associate professor, Daniel Cherqui2 MD Professor, Paolo De Simone6 MD Professor, Karim Boudjema7 MD Professor, Vincenzo Mazzaferro8 MD Professor, Olivier Soubrane9 MD Professor, Juan Carlos García-Valdecasas10 MD Professor, Joan Fabregat Prous11 MD Associate professor, Antonio D. Pinna12 MD Professor, John O’Grady MD Professor13, Vincent Karam2 MD PhD, Christophe Duvoux14 MD, Allan Rasmussen1 MD for the European Liver and Intestine Transplant Association (ELITA)
Affiliations
1: Department of Surgical Gastroenterology and Transplantation, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark
Accepted Article
2: Department of Hepatobiliary Surgery, Cancer and Transplantation, AP-HP, Hôpital Universitaire Paul Brousse, Inserm U 935, Univ Paris-Sud, Villejuif, France
3: National Institute of Public Health, University of Southern Denmark, Copenhagen, Denmark
4: Liver Transplant Center and General Surgery, A.O.U. Città della Salute e della Scienza di Torino, Molinette Hospital, Turin, Italy.
5: Department of Surgery - Liver Transplant Unit, Hospital Virgen del Rocio, Sevilla, Spain
6: Hepatobiliary Surgery and Liver Transplantation Unit, University of Pisa Medical School Hospital, 56124 Pisa, Italy.
7: Service de Chirurgie Hépatobiliaire et Digestive, Hôpital Pontchaillou, Centre Hospitalier Université de Rennes 1, Rennes, France; INSERM, UMR991, Foie, Métabolisme et Cancer, Université de Rennes 1, Rennes, France.
8: University of Milan and Division of Gastrointestinal Surgery and Liver Transplantataion, Istituto Nazionale Tumori, Fondazione IRCCS, Milan, Italy
9: Department of HPB Surgery and Liver Transplant, Beaujon Hospital, Clichy, University Denis Diderot, Paris, France
10: Hepatobiliopancreatic & Transplant Surgery, ICMDiM, Hospital Clínic, Barcelona, Spain
11: Unitat de Cirurgia Hepato-bilio-pancreàtica, Hospital Universitari de Bellvitge, Barcelona, Spain
12: General Surgery and Transplant Division, S. Orsola Hospital, University of Bologna, Bologna, Italy
13: Institute of Liver Studies, King's College Hospital, London, United Kingdom.
14: Department of Hepatology and Liver Transplant Unit Henri Mondor Hospital, Paris Est University (UPEC), Créteil, France
Accepted Article
Authorship
Hans-Christian Pommergaard: conception and design, data analysis and interpretation of results, writing the first draft, critical revision, final approval
Andreas Arendtsen Rostved: conception and design, interpretation of results, critical revision, final approval
René Adam: conception and design, acquisition of data, critical revision, final approval
Lau Caspar Thygesen: conception and design, interpretation of results, critical revision, final approval
Mauro Salizzoni: acquisition of data, critical revision, final approval
Miguel Angel Gómez Bravo: acquisition of data, critical revision, final approval
Daniel Cherqui: acquisition of data, critical revision, final approval
Paolo De Simone: acquisition of data, critical revision, final approval
Karim Boudjema: acquisition of data, critical revision, final approval
Vincenzo Mazzaferro: acquisition of data, critical revision, final approval
Olivier Soubrane: acquisition of data, critical revision, final approval
Juan Carlos García-Valdecasas: acquisition of data, critical revision, final approval
Joan Fabregat Prous: acquisition of data, critical revision, final approval
Antonio D. Pinna: acquisition of data, critical revision, final approval
John O’Grady: acquisition of data, critical revision, final approval
Vincent Karam: conception and design, acquisition of data, critical revision, final approval
Christophe Duvoux: conception and design, acquisition of data, critical revision, final approval
Accepted Article
Allan Rasmussen: conception and design, interpretation of results, acquisition of data, critical revision, final approval
Corresponding author contact information
Hans-Christian Pommergaard, MD, PhD
Department of Surgical Gastroenterology and Transplantation, Rigshospitalet, University of Copenhagen, Blegdamsvej 9, 2100 Copenhagen, Denmark, Telephone +45 23 24 18 21, Email:
HCPommergaard@gmail.com
Running title
Locoregional treatment in transplantation for hepatocellular carcinoma
Funding
ELTR received financial support from Astellas Pharma Europe Ltd., Novartis Pharma S.A.S., Institut Georges Lopez S.A.S. and ELITA (European Liver and Intestine Transplant Association). The authors of this manuscript have no conflicts of interest to disclose.
Key words
Hepatocellular carcinoma; HCC; Liver transplantation; Ltx; locoregional treatment; pretreatment
Abbreviations
Liver transplantation (Ltx)
Hepatocellular carcinoma (HCC)
Accepted Article
European Liver Transplant Registry (ELTR)
Radiofrequency ablation (RFA)
Transarterial chemoembolization (TACE)
Hazards ratio (HR)
95% confidence intervals (CI)
Model for End-Stage Liver Disease (MELD)
Conflict of interest statement
There are no conflicts of interest associated with the study for any of the authors.
Abstract
Introduction
Locoregional treatment while on the waiting list for liver transplantation (Ltx) for hepatocellular carcinoma (HCC) has been shown to improve survival. However, the effect of treatment type has not been investigated. We investigate the effect of locoregional treatment type on survival after Ltx for HCC.
Methods
We investigated patients registered in the European Liver Transplant Registry (ELTR) database using multivariate Cox regression survival analysis.
Results
Information on locoregional therapy was registered for 4,978 of 23,124 patients and was associated with improved overall survival (hazard ratio (HR) 0.84 [0.73-0.96]) and HCC-specific survival (HR 0.76
Accepted Article
[0.59-0.98]). Radiofrequency ablation (RFA) was the one monotherapy associated with improved overall survival (HR 0.51 [0.40-0.65]). In addition, the combination of RFA and transarterial chemoembolization (TACE) also improved survival (HR 0.74 [0.55-0.99]).
Conclusion
Adjusting for factors related to prognosis, disease severity and tumor aggressiveness, RFA was highly beneficial for overall and HCC-specific survival. The effect may represent a selection of patients with favorable tumor biology; however, the treatment may be effective per se by halting tumor
progression. Clinicaltrials.gov number: NCT02995096.
Introduction
In patients awaiting liver transplantation (Ltx) for hepatocellular carcinoma (HCC), disease burden may progress beyond transplantation criteria while on the waiting list. Model for End-Stage Liver Disease (MELD) exception points were introduced to alleviate dropouts due to tumor progression [1]. In addition, locoregional treatments may be used to prevent progression during waiting time [2- 4] or to downstage patients initially outside transplantation criteria [5, 6].
Response to locoregional treatment is correlated with improved recurrence-free survival after Ltx, [7] suggesting it to be a surrogate marker of tumor aggressiveness that may be used to select patients with acceptable outcome after Ltx [8]. This is supported by studies where patients downstaged from being outside transplantation criteria have similar survival to patients inside criteria [5, 6].
Accepted Article
Locoregional treatments may halt tumor progression. However, an induction of the immune system may also explain the improved prognosis [9-11]. Investigating this, smaller observational studies have not shown convincing evidence of benefit from locoregional treatments [12-16]. Conversely, intention-to-treat studies following patients from listing and including dropouts suggested a benefit [5, 6].
Whether locoregional treatments are beneficial or merely mirror selection of patients with favorable tumor biology has not been fully elucidated. While the present study, with only data on transplanted patients, does not account for dropouts, the large sample size representing general clinical practice may elucidate important factors such as the importance of the type of locoregional treatment.
The aim was to investigate the impact of locoregional treatment including types of treatment before liver transplantation for HCC on survival and HCC-specific survival in a large cohort from the
European Liver Transplant Registry (ELTR).
Methods
Reporting of this study complied with the guidelines laid out in the STROBE statement [17]. Prior to performing data analyses, a protocol was registered at clinicaltrials.gov (ID NCT02995096). This was a retrospective cohort study using register-based data recorded in the ELTR database. This database is comprised of information from 172 liver transplantation centers across Europe, each reporting local data pre- and post-transplantation in patients undergoing Ltx. No data were recorded for patients who dropped out from the waiting list. The database contains information on donor, recipient, locoregional treatments before transplantation, immunosuppression, pathology from the explanted liver, underlying liver disease, presence of cirrhosis in addition to HCC, time of death, and cause of death. Data are electronically reported to the central ELTR database from each center.
Accepted Article
We included patients registered for transplant in ELTR from 1990 to November 2016 due to HCC.
Participants with information on locoregional treatments were included. Locoregional treatment was defined as localized treatment for the HCC tumor(s) in order to downstage or prevent progression outside criteria while on the waiting list.
The outcomes were five-year overall survival and HCC-specific survival after transplantation.
Exposures were locoregional treatment (yes/no), locoregional treatment types (radiofrequency ablation (RFA) as monotherapy, transarterial chemoembolization (TACE) as monotherapy, resection as monotherapy, other treatment as monotherapy, RFA + TACE, RFA + TACE + other, other
combinations) and number of locoregional treatments (0, 1, 2, 3 or more)
We used a multiple imputation model with fully conditional specification and five imputations due to missing data on covariates. The model included number of nodules, size of largest nodule, vascular invasion, time on waiting list, cirrhosis, age, gender, and MELD-score. Before and after imputation, the distribution and mean values for variables were comparable. All analyses were repeated with 50 imputations instead of five, which did not change the magnitude of the estimates.
Univariate Cox regression models were used to evaluate association between locoregional
treatment and five-year survival as well as five-year HCC-specific survival from the time of Ltx. These were reported as hazards ratios (HR) and 95% confidence intervals (CI). Outcome data and follow-up time for these analyses were censored after five years. To test the independent effect of
locoregional treatment, multivariate models were done for each exposure variable and adjusted for the plausible confounders of gender, age, time on waiting list, number of nodules, maximum size of nodules, vascular invasion (micro, macro or none), cirrhosis and MELD-score. We checked the proportional hazards assumption for covariates with log-minus-log plots with the natural logarithm of follow-up time. For the continuous variables (time on waiting list, number of nodules, maximum size of nodules, MELD-score), the linear effect was evaluated by including the second order
polynomial of the variables in the model. No deviation from linearity was found. To test for a change
Accepted Article
in effect over time, we included an interaction term between the locoregional treatment yes/no variable and a categorical variable with four periods (1990-1996, 1997-2003, 2004-2009 and 2010- 2016) in the multivariate COX regression model. Five-year overall and HCC-specific survival were analyzed as cumulative survival with a 95% CI using Kaplan Meier statistics, and groups were compared with a log-rank test. The Mann-Whitney U-test was used to analyze continuous data that was not normally distributed, and a Chi-square test or Fisher’s exact test was used to evaluate nominal data. Using the Reserve-Kaplan Meier method, we reported the median follow-up time with interquartile range (IQR) [18]. We used IBM SPSS Statistics version 23, with statistical significance defined as p≤0.05.
Results
Patient characteristics (Table 1)
From 23,124 patients transplanted for HCC in the ELTR, 4,978 patient records had data on locoregional treatments. Patients excluded due to missing data on locoregional treatments were comparable to the included patients with respect to age, gender, number of nodules, size of largest nodule, vascular invasion, MELD-score and cirrhosis (Supplementary Table 1). The majority (71.8%) of the included patients received locoregional treatment and 85.2% of these received just one treatment. Only 1.5% received three or more treatments. TACE was most common treatment (59.1%), followed by RFA (18.0%).
Time on waiting list was longer for patients receiving locoregional treatment (median 118 vs. 49 days, p<0.001, Mann-Whitney U-test). This difference remained significant when stratified by size of largest nodule (> 5 cm and ≤ 5 cm). In addition, fewer patients had vascular invasion (p=0.003, Fisher’s exact test), and the average MELD-score was lower in the locoregional treatment group
Accepted Article
(median 10.1 vs 12.2, p>0.001 Mann-Whitney). However, the groups were comparable with regards to being inside the Milan criteria (40.6 vs. 42.7%, p=0.20, Chi-square test).
Locoregional treatment (Table 2 and 3)
The five-year overall survival rate was 69.7% [67.7-71.7] for patients receiving locoregional
treatment compared with 65.8% [62.5-69.1] for patients not receiving treatment (p< 0.001, Log-rank test). Locoregional treatment was significantly associated with improved prognosis for five-year overall and HCC-specific survival in univariate analyses and the estimates remained largely unchanged and significant in multivariate analyses (Tables 2 and 3). We found no significant interaction between the year of transplantation and locoregional treatment, suggesting that the effects were comparable throughout the study period.
Locoregional treatment types
The five-year overall survival rate was 80.9% [77.3-84.7] for patients treated with RFA, 67.6% [65.1- 70.2] for TACE treatment and 51.3% [40.5-62.1] for resection compared with 65.8% [62.5-69.1] for patients not receiving locoregional treatment. Regarding treatment type (Tables 2 and 3), RFA had the strongest association with improved overall survival, both in univariate and multivariate analyses (HR 0.51 [0.40-0.65]). The effect was even stronger for HCC-specific survival (HR 0.43 [0.26-0.69]). In contrast, TACE was not significant in multivariate analysis (HR 0.89 [0.77-1.03]). However, the combination of RFA and TACE also improved survival (HR 0.74 [0.55-0.99]). Conversely, resection was associated with reduced overall survival (HR 1.37 [1.02-1.83]). The remaining treatments were not significant for five-year overall survival in multivariate analysis.
Accepted Article
The differences in survival between types of locoregional treatment are shown in Figure 1. The median follow-up time was 26 months (IQR, 7-60 months) with no treatment; 44 months (IQR 16-60 months) with RFA; 38 months (IQR, 13-60 months) with TACE, 35 months (IQR, 12-60 months) with resection; and 33 months (IQR, 11-60 months) for other treatments, including combination treatments.
Number of treatments
In univariate analysis, two treatments showed stronger association with improved survival than one treatment. However, three or more treatments showed no association. These estimates remained unchanged in multivariate analysis (Tables 2 and 3).
Subgroup analyses
More than five nodules or size of largest nodule > 3 cm
Patients with more than five nodules, or whose largest nodule was > 3 cm were evaluated in a subgroup analysis, since RFA treatment is normally not considered suitable for these patients. In a multivariate model of overall five-year survival, locoregional treatment remained significant (0.78 [0.65-0.94]). Evaluated based on type of treatment, RFA remained significantly associated with improved survival, with a HR of 0.54 [0.39-0.77]. However, TACE was now also significantly associated with improved survival (HR 0.81 [0.67-0.98]). Furthermore, the combination of RFA and TACE was associated with improved survival (HR 0.60 [0.39-0.93]). There was no association for resected patients (HR 1.05 [0.69-1-60]).
Accepted Article
Cirrhosis vs. non-cirrhosis
The distribution of types of locoregional treatment was significantly different between patients with and without cirrhosis (p<0.001, Chi-square test). More patients with cirrhosis received TACE alone (43.0 vs. 29.4%), fewer patients with cirrhosis received resection (3.2 vs. 7.2%), and fewer patients with cirrhosis received RFA alone (12.2 vs. 28.1%). In a multivariate model for overall five-year survival for the subgroup of patients with cirrhosis, the effect of locoregional treatment (HR 0.86 [0.75-0.99]) was comparable to that of the whole group. Similarly, RFA was associated with improved survival (HR 0.52 [0.40-0.67]), TACE was not (HR 0.91 [0.78-1.05]), and resection was associated with diminished survival (HR 1.48 [1.09-2.00]).
Vascular invasion
The effect of locoregional treatment was stronger for patients with micro- or macrovascular invasion (HR 0.71 [0.55-0.92]) compared to patients with no vascular invasion, where no effect was observed (HR 0.88 [0.75-1.05]). Regarding type of treatment, both RFA (HR 0.54 [0.32-0.91]) and TACE (HR 0.69 [0.52-0.92]) were associated with improved survival for patients with micro- or macrovascular invasion.
Discussion
Adjusted for disease severity, the present study showed that locoregional treatment with RFA was associated with improved overall and disease-specific survival. To a lesser extent, the combination of RFA and TACE was also beneficial. Subgroup analyses showed that the effect was stronger for patients with vascular invasion. Moreover, patients with extensive disease still experienced improved outcomes from RFA.
Accepted Article
Previous studies have shown that locoregional treatment may successfully downstage patients to be inside Milan criteria and improve survival following liver transplantation [5, 6, 16, 19, 20]. Two intention-to-treat studies have been conducted in which patients were followed since listing for transplantation [5, 6]. In one study, 118 patients scheduled for downstaging treatment with TACE/
RFA were compared with 488 patients within the Milan T2 criteria at the time of listing [5].
Downstaging was successful in 65.3% of the cases and 54.2% received Ltx. The five-year survival rate was 56.1% for all patients receiving downstaging treatment vs. 63.3% for patients meeting the T2 Milan criteria (p=0.29). Among the patients proceeding to Ltx, the five-year survival rate was 77.8%
in the downstaging groups and 81% in the Milan T2 group (p=0.69). In 40.6% of patients receiving downstaging, there was no residual tumor in explant pathology. In another study, 129 patients inside Milan criteria were compared with 48 patients outside the criteria receiving downstaging treatment; the results showed comparable transplantation rates and intention-to-treat survival between the groups [6].
A direct effect of the locoregional treatments may be induction of the immune system to produce a response toward the tumor, causing a reduction in tumor size, as well as halting progression. In one study, RFA induced stimulation of natural killer cells, demonstrating a benefit to recurrence-free survival among patients with the largest response [11]. Moreover, in another study, the antigen- specific response from T-cells was enhanced in 31 patients with HCC receiving locoregional treatments [10]. Lastly, at least one study has shown that the extent of necrosis induced by
locoregional treatment was associated with cell-mediated tumor rejection and clinical response [9].
In a clinical study, 101 patients randomized to TACE or transarterial embolization (TAE) had
comparable response, progression free survival and overall survival. Thus, questioning the additional benefit of local chemotherapy for these patients besides the effect induces by ischemia [21].
Accepted Article
Surprisingly, patients in a downstaging study had extremely low rates of microvascular invasion in explant pathology (1.6% versus 18.6% for all patients receiving locoregional treatment in the present study) [5]. Moreover, in the present study, vascular invasion was less common among patients receiving locoregional treatment. Thus, while intention-to-treat studies suggest an effect of locoregional treatments per se [5, 6], these treatments may also select patients with favorable tumor biology. Patients with progression-free waiting time after TACE had a lower risk of recurrence, possibly related to less aggressive tumors [7]. This may be illustrated by the longer time on the waiting list for patients receiving locoregional treatment in the present study. Thus, one may argue that transplantation should be limited to patients without progression after locoregional treatment.
In fact, selection on this basis may be superior to selecting based on the Milan criteria [7], as those guidelines may exclude patients outside criteria with favorable tumor biology. This is supported by comparable outcomes between downstaged patients and patients inside Milan criteria [5, 6].
Correspondingly, we show that an indirect measure of tumor aggressiveness (vascular invasion) may be a superior prognostic marker compared with the Milan criteria [22].
Our results suggest that resection was associated with reduced overall survival compared with no treatment. Randomized controlled trials comparing RFA and resection as primary treatment for patients with small HCC [23] and HCC with Milan criteria [24] agree that resection is the superior treatment. Therefore, despite adjusting for tumor characteristics, our results may be a result of selection bias, where patients with large tumors, and thus a worse prognosis, are limited to surgical resection.
This is the first study to compare RFA, TACE and other locoregional treatment types on a large scale, and to show that RFA may be superior. It is a pan-European study, with many centers reflecting general clinical practice with a broad external validity. An important limitation is that we could not account for patients dropping out from the waiting list; thus, only transplanted patients were evaluated. While approximately 7% will drop out while on the waiting list due to tumor progression
Accepted Article
[4], the included patients may represent a selective sample, presumably including more patients with favorable tumor biology. With longer time on waiting list and less vascular invasion, the
locoregional treatment group may have a higher proportion of patients with favorable tumor biology and, thus, good response to treatment or stable disease, allowing subsequent transplantation.
Unfortunately, response to locoregional treatment was not included in the ELTR, posing another limitation. With a considerable amount of missing data, we had information on locoregional treatments for only 4,978 of 23,124 patients transplanted for HCC in the study period. This was a result of locoregional data not being included in the ELTR questionnaire before 2007-2008. Thus, these data may not reflect the true picture of the whole population; however, the two populations were comparable with respect to prognostic variables. Lastly, despite adjusting for indirect measures of tumor aggressiveness and tumor load (size, number and vascular invasion), the benefit from RFA may be a product of residual confounding resulting from patients treated with RFA having less severe disease prior to treatment.
Studies have shown that response to locoregional treatment is an important factors for prognosis [7, 25]. Thus, as proposed by Mazzaferro [8], future selection criteria may be based on response to locoregional treatments. In other words, all patients with HCC on a waiting list for Ltx should be treated with locoregional treatments followed by a suitable observation period regardless of tumor size and number. Transplant priority should be given to patients based not only on conventional criteria, but also on the response to locoregional treatment [8]. The results of the present paper support this notion.
In conclusion, we show that locoregional treatments with RFA were beneficial for both overall and HCC-specific survival. After adjusting for variables related to tumor aggressiveness, treatment with RFA was still highly effective. These results may represent a selection of patients with favorable tumor biology who responded to locoregional treatments and subsequently received Ltx; however, intention-to-treat studies suggest an effect of locoregional treatments per se. Thus, induction of the
Accepted Article
immune system by locoregional treatments, rather than, or in addition to, patient selection, may explain these observations.
Figure legends
Figure 1: Overall survival for different types of locoregional treatment RFA: Radiofrequency ablation, TACE: Transarterial chemoembolization
References
1. Wiesner RH, Freeman RB, Mulligan DC. Liver transplantation for hepatocellular cancer: the impact of the MELD allocation policy. Gastroenterology. 2004; 127: S261-7.
2. Fontana RJ, Hamidullah H, Nghiem H, Greenson JK, Hussain H, Marrero J, et al. Percutaneous radiofrequency thermal ablation of hepatocellular carcinoma: a safe and effective bridge to liver transplantation. Liver Transpl. 2002; 8: 1165-74.
3. Harnois DM, Steers J, Andrews JC, Rubin JC, Pitot HC, Burgart L, et al. Preoperative hepatic artery chemoembolization followed by orthotopic liver transplantation for hepatocellular carcinoma. Liver Transpl Surg. 1999; 5: 192-9.
4. Lee MW, Raman SS, Asvadi NH, Siripongsakun S, Hicks RM, Chen J, et al. Radiofrequency Ablation of Hepatocellular Carcinoma as Bridge Therapy to Liver Transplantation: A Ten Year Intention-to-treat Analysis. Hepatology. 2017.
5. Yao FY, Mehta N, Flemming J, Dodge J, Hameed B, Fix O, et al. Downstaging of hepatocellular cancer before liver transplant: long-term outcome compared to tumors within Milan criteria.
Hepatology. 2015; 61: 1968-77.
6. Ravaioli M, Grazi GL, Piscaglia F, Trevisani F, Cescon M, Ercolani G, et al. Liver transplantation for hepatocellular carcinoma: results of down-staging in patients initially outside the Milan selection criteria. Am J Transplant. 2008; 8: 2547-57.
7. Otto G, Herber S, Heise M, Lohse AW, Monch C, Bittinger F, et al. Response to transarterial chemoembolization as a biological selection criterion for liver transplantation in
hepatocellular carcinoma. Liver Transpl. 2006; 12: 1260-7.
8. Mazzaferro V. Squaring the circle of selection and allocation in liver transplantation for HCC:
An adaptive approach. Hepatology. 2016; 63: 1707-17.
9. Ayaru L, Pereira SP, Alisa A, Pathan AA, Williams R, Davidson B, et al. Unmasking of alpha- fetoprotein-specific CD4(+) T cell responses in hepatocellular carcinoma patients undergoing embolization. J Immunol. 2007; 178: 1914-22.
10. Mizukoshi E, Nakamoto Y, Arai K, Yamashita T, Sakai A, Sakai Y, et al. Comparative analysis of various tumor-associated antigen-specific t-cell responses in patients with hepatocellular carcinoma. Hepatology. 2011; 53: 1206-16.
11. Zerbini A, Pilli M, Laccabue D, Pelosi G, Molinari A, Negri E, et al. Radiofrequency thermal ablation for hepatocellular carcinoma stimulates autologous NK-cell response.
Gastroenterology. 2010; 138: 1931-42.
12. Yao FY, Ferrell L, Bass NM, Watson JJ, Bacchetti P, Venook A, et al. Liver transplantation for hepatocellular carcinoma: expansion of the tumor size limits does not adversely impact survival. Hepatology. 2001; 33: 1394-403.
Accepted Article
13. Regalia E, Coppa J, Pulvirenti A, Romito R, Schiavo M, Burgoa L, et al. Liver transplantation for small hepatocellular carcinoma in cirrhosis: analysis of our experience. Transplant Proc.
2001; 33: 1442-4.
14. Oldhafer KJ, Chavan A, Fruhauf NR, Flemming P, Schlitt HJ, Kubicka S, et al. Arterial
chemoembolization before liver transplantation in patients with hepatocellular carcinoma:
marked tumor necrosis, but no survival benefit? J Hepatol. 1998; 29: 953-9.
15. Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A, Bozzetti F, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med. 1996; 334: 693-9.
16. Yao FY, Kinkhabwala M, LaBerge JM, Bass NM, Brown R, Jr., Kerlan R, et al. The impact of pre-operative loco-regional therapy on outcome after liver transplantation for
hepatocellular carcinoma. Am J Transplant. 2005; 5: 795-804.
17. von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement:
guidelines for reporting observational studies. J Clin Epidemiol. 2008; 61: 344-9.
18. Schemper M, Smith TL. A note on quantifying follow-up in studies of failure time. Control Clin Trials. 1996; 17: 343-6.
19. Majno PE, Adam R, Bismuth H, Castaing D, Ariche A, Krissat J, et al. Influence of preoperative transarterial lipiodol chemoembolization on resection and transplantation for hepatocellular carcinoma in patients with cirrhosis. Ann Surg. 1997; 226: 688-701; discussion -3.
20. De Luna W, Sze DY, Ahmed A, Ha BY, Ayoub W, Keeffe EB, et al. Transarterial chemoinfusion for hepatocellular carcinoma as downstaging therapy and a bridge toward liver
transplantation. Am J Transplant. 2009; 9: 1158-68.
21. Brown KT, Do RK, Gonen M, Covey AM, Getrajdman GI, Sofocleous CT, et al. Randomized Trial of Hepatic Artery Embolization for Hepatocellular Carcinoma Using Doxorubicin-Eluting Microspheres Compared With Embolization With Microspheres Alone. J Clin Oncol. 2016; 34:
2046-53.
22. Pommergaard HC, Rostved AA, Adam R, Thygesen LC, Salizzoni M, Bravo M, et al. Vascular invasion and survival after liver transplantation for hepatocellular carcinoma: A study from the European Liver Transplant Registry Transplantation (Submitted for publication). 2017.
23. Feng K, Yan J, Li X, Xia F, Ma K, Wang S, et al. A randomized controlled trial of radiofrequency ablation and surgical resection in the treatment of small hepatocellular carcinoma. J
Hepatol. 2012; 57: 794-802.
24. Huang J, Yan L, Cheng Z, Wu H, Du L, Wang J, et al. A randomized trial comparing
radiofrequency ablation and surgical resection for HCC conforming to the Milan criteria. Ann Surg. 2010; 252: 903-12.
25. Allard MA, Sebagh M, Ruiz A, Guettier C, Paule B, Vibert E, et al. Does pathological response after transarterial chemoembolization for hepatocellular carcinoma in cirrhotic patients with cirrhosis predict outcome after liver resection or transplantation? J Hepatol. 2015; 63: 83-92.
Accepted Article
Patient characteristics and missing data
No locoregional treatment (n=1406)Locoregional treatment (n=
3572)
Missing data, n (%)
Recipient age, median (range)
55 (0-78) 58 (0-77) 2 (0%)
Male gender, n (%) 1151 (81.9%) 3030 (84.8%) 1 (0%)
AB0 matching, n (%) 75 (1.5%)
Identical 1185 (85.8%) 3305 (93.8%)
Compatible 177 (12.8%) 188 (5.3%)
Non-compatible 19 (1.4%) 29 (0.8%)
Number of
transplantations, n (%)
0 (0%)
1 1354 (96.3%) 3378 (94.6%)
2 47 (3.3%) 186 (5.2%)
3 or more 5 (0.4%) 8 (0.2%)
Number of locoregional treatments, n (%)
0 (0%)
1 3042 (85.2%)
2 477 (13.4%)
3 or more 53 (1.5%)
Type of locoregional treatment, n (%)
0 (0%)
RFA as monotherapy 643 (18%)
TACE as monotherapy 2110 (59.1%)
Resection as monotherapy 169 (4.7%)
Other as monotherapy 120 (3.4%)
RFA+ TACE 280 (7.8%)
RFA + TACE + other 49 (1.4%)
Other combination 201 (5.6%)
Accepted Article
Days on waitinglist, median (range)
49 (0-8492) 118 (0-3991) 777 (15.6%)
Transplanted after 2006, n (%)
1126 (80.1%) 2949 (82.6%) 0 (0%)
Outside Milan criteria, n (%)
531 (42.7%) 1143 (40.6%) 919 (18.5%)
More than 3 nodules, n (%)
197 (14.9%) 561 (17.6%) 471 (9.5%)
Size of largest nodules > 5 cm, n (%)
230 (17.8%) 328 (10.6%) 590 (11.9%)
Vascular invasion, n (%) 849 (17.1%)
None 933 (74.9%) 2264 (78.5%)
Macrovascular 59 (4.7%) 83 (2.9%)
Microvascular 254 (20.4%) 536 (18.6%) MELD score, median
(range)
12.1 (6.4-49.6) 10.1 (6.4-42.8) 1519 (30.5%)
Cirrose, n (%) 0 (0%)
Non-cirrose 42 (3%) 179 (5%)
Cirrose 1358 (96.6%) 3385 (94.8%)
Fibrolamellar 6 (0.4%) 8 (0.2%)
Accepted Article
Overall 5 year survival Hazard ratio [95% CI]
HCC specific 5 year survival
Hazard ratio [95% CI]
HR [CI 95%] HR [CI 95%]
Age (10 year increase) 1.10 [1.02-1.18] 0.92 [0.82-1.04]
Gender (Male) 0.90 [0.77-1.06] 0.88 [0.78-1.00]
Time on waiting list (30 days increase)
1.00 [1.00-1.01] 1.00 [0.99-1.01]
More than 3 nodules 1.61 [1.51-1.71] 2.59 [2.33-2.87]
Size of largest nodule > 5 cm 2.00 [1.88-2.14] 2.99 [2.68-3.33]
Vascular invasion
No vascular invasion Ref. Ref.
Macrovascular invasion 2.66 [2.04-3.47] 8.87 [7.64-10.30]
Microvascular invasion 1.57 [1.36-1.82] 3.16 [2.86-3.49]
MELD-score 1.04 [1.03-1.05] 1.01 [1.00-1.02]
Cirrose
Non-cirrosis Ref. Ref.
Cirrosis 0.86 [0.67-1.11] 0.59 [0.50-0.69]
Fibrolamellar 2.10 [0.91-4.86] 3.02 [1.86-4.92]
Locoregional treatment (yes/no) 0.82 [0.72-0.94] 0.73 [0.66-0.80]
Locoregional treatments
No treatment Ref. Ref.
RFA as monotherapy 0.47 [0.37-0.60] 0.35 [0.29-0.43]
TACE as monotherapy 0.90 [0.78-1.03] 0.78 [0.70-0.86]
Resection as monotherapy 1.34 [1.00-1.78] 1.24 [0.99-1.55]
Other as monotherapy 1.19 [0.85-1.67] 1.36 [1.07-1.72]
RFA+ TACE 0.76 [0.57-1.02] 0.74 [0.60-0.92]
RFA + TACE + other 1.03 [0.58-1.82] 1.45[1.00-2.09]
Accepted Article
Univariate COX-regression analyses
Number of locoregional treatments0 Ref. Ref.
1 0.83 [0.73-0.95] 0.73 [0.65-0.80]
2 0.72 [0.57-0.91] 0.70 [0.58-0.84]
3 or more 1.17 [0.69-2.00] 1.44 [1.00-2.08]
Accepted Article
Multivariate COX-regression analysis
Overall 5 year survival Hazard ratio [95% CI]
HCC specific 5 year survival
Hazard ratio [95%
CI]
Locoregional treatment (yes/no) 0.84 [0.73-0.96] 0.76 [0.59-0.98]
Locoregional treatments
No treatment Ref. Ref.
RFA as monotherapy 0.51 [0.40-0.65] 0.43 [0.26-0.69]
TACE as monotherapy 0.89 [0.77-1.03] 0.79 [0.60-1.04]
Resection as monotherapy 1.37 [1.02-1.83] 1.22 [0.70-2.14]
Other as monotherapy 1.20 [0.85-1.69] 1.50 [0.83-2.70]
RFA+ TACE 0.74 [0.55-0.99] 0.65 [0.38-1.11]
RFA + TACE + other 0.98 [0.55-1.75] 1.35 [0.54-3.35]
Other combination 0.71 [0.49-1.02] 0.66 [0.33-1.33]
Number of locoregional treatments
0 Ref. Ref.
1 0.85 [0.74-0.98] 0.77 [0.59-1.00]
2 0.71 [0.56-0.91] 0.66 [0.42-1.03]
3 or more 1.11 [0.65-1.91] 1.35 [0.54-3.35]
Accepted Article
Figure 1: Overall survival for types of locoregional treatment
Follow-up time 20 months 40 months 60 months
No treatment Numbers at risk 626 413 290
Survival (%) 78.6 71.2 65.8
RFA alone Numbers at risk 417 302 199
Survival (%) 90.1 85.2 80.9
TACE alone Numbers at risk 1153 743 500
Survival (%) 80.5 72.4 67.6
Resection alone Numbers at risk 84 50 28
Survival (%) 74.4 65.4 51.3
Other including combination treatments
Numbers at risk 427 257 175
Survival (%) 80.3 72.7 65.5
RFA: Radiofrequency ablation, TACE: Transarterial chemoembolization
